virtualspecies,an R package to generate virtual species distributions

virtualspecies is a freely available package for R designed to generate virtual species distributions, a procedure increasingly used in ecology to improve species distribution models. This package combines the existing methodological approaches with the objective of generating virtual species distributions with increased ecological realism. The package includes 1) generating the probability of occurrence of a virtual species from a spatial set of environmental conditions (i.e. environmental suitability), with two different approaches; 2) converting the environmental suitability into presence–absence with a probabilistic approach; 3) introducing dispersal limitations in the realised virtual species distributions and 4) sampling occurrences with different biases in the sampling procedure. The package was designed to be extremely flexible, to allow users to simulate their own defined species–environment relationships, as well as to provide a fine control over every simulation parameter. The package also includes a function to generate random virtual species distributions. We provide a simple example in this paper showing how increasing ecological realism of the virtual species impacts the predictive performance of species distribution models. We expect that this new package will be valuable to researchers willing to test techniques and protocols of species distribution models as well as various biogeographical hypotheses.     

ecospat: an R package to support spatial analyses and modeling of species niches and distributions

The aim of the ecospat package is to make available novel tools and methods to support spatial analyses and modeling of species niches and distributions in a coherent workflow. The package is written in the R language (R Development Core Team) and contains several features, unique in their implementation, that are complementary to other existing R packages. Pre‐modeling analyses include species niche quantifications and comparisons between distinct ranges or time periods, measures of phylogenetic diversity, and other data exploration functionalities (e.g. extrapolation detection, ExDet). Core modeling brings together the new approach of ensemble of small models (ESM) and various implementations of the spatially‐explicit modeling of species assemblages (SESAM) framework. Post‐modeling analyses include evaluation of species predictions based on presence‐only data (Boyce index) and of community predictions, phylogenetic diversity and environmentally‐constrained species co‐occurrences analyses. The ecospat package also provides some functions to supplement the ‘biomod2’ package (e.g. data preparation, permutation tests and cross‐validation of model predictive power). With this novel package, we intend to stimulate the use of comprehensive approaches in spatial modelling of species and community distributions.     

BIOMOD – a platform for ensemble forecasting of species distributions

BIOMOD is a computer platform for ensemble forecasting of species distributions, enabling the treatment of a range of methodological uncertainties in models and the examination of species-environment relationships. BIOMOD includes the ability to model species distributions with several techniques, test models with a wide range of approaches, project species distributions into different environmental conditions (e.g. climate or land use change scenarios) and dispersal functions. It allows assessing species temporal turnover, plot species response curves, and test the strength of species interactions with predictor variables. BIOMOD is implemented in R and is a freeware, open source, package.     

MigClim: Predicting plant distribution and dispersal in a changing climate

Aim Many studies have forecasted the possible impact of climate change on plant distributions using models based on ecological niche theory, but most of them have ignored dispersal‐limitations, assuming dispersal to be either unlimited or null. Depending on the rate of climatic change, the landscape fragmentation and the dispersal capabilities of individual species, these assumptions are likely to prove inaccurate, leading to under‐ or overestimation of future species distributions and yielding large uncertainty between these two extremes. As a result, the concepts of ‘potentially suitable’ and ‘potentially colonizable’ habitat are expected to differ significantly. To quantify to what extent these two concepts can differ, we developed Mig Clim, a model simulating plant dispersal under climate change and landscape fragmentation scenarios. Mig Clim implements various parameters, such as dispersal distance, increase in reproductive potential over time, landscape fragmentation or long‐distance dispersal. Location Western Swiss Alps. Methods Using our Mig Clim model, several simulations were run for two virtual species by varying dispersal distance and other parameters. Each simulation covered the 100‐year period 2001–2100 and three different IPCC‐based temperature warming scenarios were considered. Results of dispersal‐limited projections were compared with unlimited and no‐dispersal projections. Results Our simulations indicate that: (1) using realistic parameter values, the future potential distributions generated using Mig Clim can differ significantly (up to more than 95% difference in colonized surface) from those that ignore dispersal; (2) this divergence increases under more extreme climate warming scenarios and over longer time periods; and (3) the uncertainty associated with the warming scenario can be as large as the one related to dispersal parameters. Main conclusions Accounting for dispersal, even roughly, can importantly reduce uncertainty in projections of species distribution under climate change scenarios.     

Environmental conditions affect spatial genetic structures and dispersal patterns in a solitary rodent

The study of the spatial distribution of relatives in a population under contrasted environmental conditions provides critical insights into the flexibility of dispersal behaviour and the role of environmental conditions in shaping population relatedness and social structure. Yet few studies have evaluated the effects of fluctuating environmental conditions on relatedness structure of solitary species in the wild. The aim of this study was to determine the impact of interannual variations in environmental conditions on the spatial distribution of relatives [spatial genetic structure (SGS)] and dispersal patterns of a wild population of eastern chipmunks (Tamias striatus), a solitary rodent of North America. Eastern chipmunks depend on the seed of masting trees for reproduction and survival. Here, we combined the analysis of the SGS of adults with direct estimates of juvenile dispersal distance during six contrasted years with different dispersal seasons, population sizes and seed production. We found that environmental conditions influences the dispersal distances of juveniles and that male juveniles dispersed farther than females. The extent of the SGS of adult females varied between years and matched the variation in environmental conditions. In contrast, the SGS of males did not vary between years. We also found a difference in SGS between males and females that was consistent with male‐biased dispersal. This study suggests that both the dispersal behaviour and the relatedness structure in a population of a solitary species can be relatively labile and change according to environmental conditions.     

Matching habitat choice promotes species persistence under climate change

Species may survive under contemporary climate change by either shifting their range or adapting locally to the warmer conditions. Theoretical and empirical studies recently underlined that dispersal, the central mechanism behind these responses, may depend on the match between an individuals’ phenotype and local environment. Such matching habitat choice is expected to induce an adaptive gene flow, but it now remains to be studied whether this local process could promote species’ responses to climate change. Here, we investigate this by developing an individual‐based model including either random dispersal or temperature‐dependent matching habitat choice. We monitored population composition and distribution through space and time under climate change. Relative to random dispersal, matching habitat choice induced an adaptive gene flow that lessened spatial range loss during climate warming by improving populations’ viability within the range (i.e. limiting range fragmentation) and by facilitating colonization of new habitats at the cold margin. The model even predicted range contraction under random dispersal but range expansion under optimal matching habitat choice. These benefits of matching habitat choice for population persistence mostly resulted from adaptive immigration decision and were greater for populations with larger dispersal distance and higher emigration probability. We also found that environmental stochasticity resulted in suboptimal matching habitat choice, decreasing the benefits of this dispersal mode under climate change. However population persistence was still better under suboptimal matching habitat choice than under random dispersal. Our results highlight the urgent need to implement more realistic mechanisms of dispersal such as matching habitat choice into models predicting the impacts of ongoing climate change on biodiversity.     

Predicted declines in suitable habitat for greater one‐horned rhinoceros (Rhinoceros unicornis) under future climate and land use change scenarios

Rapidly changing climate is likely to modify the spatial distribution of both flora and fauna. Land use change continues to alter the availability and quality of habitat and further intensifies the effects of climate change on wildlife species. We used an ensemble modeling approach to predict changes in habitat suitability for an iconic wildlife species, greater one‐horned rhinoceros due to the combined effects of climate and land use changes. We compiled an extensive database on current rhinoceros distribution and selected nine ecologically meaningful environmental variables for developing ensemble models of habitat suitability using ten different species distribution modeling algorithms in the BIOMOD2 R package; and we did this under current climatic conditions and then projected them onto two possible climate change scenarios (SSP1‐2.6 and SSP5‐8.5) and two different time frames (2050 and 2070). Out of ten algorithms, random forest performed the best, and five environmental variables—distance from grasslands, mean temperature of driest quarter, distance from wetlands, annual precipitation, and slope, contributed the most in the model. The ensemble model estimated the current suitable habitat of rhinoceros to be 2610 km², about 1.77% of the total area of Nepal. The future habitat suitability under the lowest and highest emission scenarios was estimated to be: (1) 2325 and 1904 km² in 2050; and (2) 2287 and 1686 km² in 2070, respectively. Our results suggest that over one‐third of the current rhinoceros habitat would become unsuitable within a period of 50 years, with the predicted declines being influenced to a greater degree by climatic changes than land use changes. We have recommended several measures to moderate these impacts, including relocation of the proposed Nijgad International Airport given that a considerable portion of potential rhinoceros habitat will be lost if the airport is constructed on the currently proposed site.     

SpatialDemography: a spatially explicit,stage‐structured,metacommunity model

The responses of species and populations to changes in the environment (e.g. changes in climate and land use) are often complex and difficult to predict. We have created the SpatialDemography model (R package: spatialdemography). The model is a spatially explicit, stage‐structured, matrix‐based metacommunity model, with the potential for modeling species’ and populations’ potential responses to environmental heterogeneity and change. The SpatialDemography model assumes a cellular landscape populated by organisms with four life stages: a mobile dispersing stage, two sessile non‐reproductive stages, and a reproductive adult stage. Individuals are assumed to originate at the center of a given cell and disperse according to a specified dispersal kernel (e.g. log‐normal). All adult individuals are capable of producing offspring. The model approach and framework are described in the context of a hypothetical example with multiple competing species in a four cell landscape. In this example simulation, both spatial location and species interactions were important for understanding population dynamics. SpatialDemography can be applied to questions where an understanding of transient and long‐term demographic responses to spatiotemporal changes is desired. It is primarily applicable to metapopulations and metacommunities of organisms with early dispersal and sessile adults (i.e. modular organisms such as plants and some marine organisms). SpatialDemography differs from other population models in that it is spatially explicit, can incorporate biotic interactions, and is implemented in R.     

Accommodating Species Climate-Forced Dispersal and Uncertainties in Spatial Conservation Planning

Spatial conservation prioritization should seek to anticipate climate change impacts on biodiversity and to mitigate these impacts through the development of dynamic conservation plans. Here, we defined spatial priorities for the conservation of amphibians inhabiting the Atlantic Forest Biodiversity Hotspot that overcome the likely impacts of climate change on the distribution of this imperiled fauna. First, we built ecological niche models (ENMs) for 431 amphibian species both for current time and for the mid-point of a 30-year period spanning 2071–2099 (i.e. 2080). For modeling species'' niches, we combined six modeling methods and three different climate models. We also quantified and mapped model uncertainties. Our consensus models forecasted range shifts that culminate with high species richness in central and eastern Atlantic Forest, both for current time and for 2080. Most species had a significant range contraction (up to 72%) and 12% of species were projected to be regionally extinct. Most species would need to disperse because suitable climatic sites will change. Therefore, we identified a network of priority sites for conservation that minimizes the distance a given species would need to disperse because of changes in future habitat suitability (i.e. climate-forced dispersal) as well as uncertainties associated to ENMs. This network also maximized complementary species representation across currently established protected areas. Priority sites already include possible dispersal corridors linking current and future suitable habitats for amphibians. Although we used the a top-ranked Biodiversity Hotspot and amphibians as a case study for illustrating our approach, our study may help developing more effective conservation strategies under climate change, especially when applied at different spatial scales, geographic regions, and taxonomic groups.     

Predicting future distributions of mountain plants under climate change: does dispersal capacity matter?

Many studies have investigated the potential impacts of climate change on the distribution of plant species, but few have attempted to constrain projections through plant dispersal limitations. Instead, most studies published so far have simplified dispersal as either unlimited or null. However, depending on the dispersal capacity of a species, landscape fragmentation, and the rate of climatic change, these assumptions can lead to serious over- or underestimation of the future distribution of plant species.
To quantify the discrepancies between simulations accounting for dispersal or not, we carried out projections of future distribution over the 21st century for 287 mountain plant species in a study area of the western Swiss Alps. For each species, simulations were run for four dispersal scenarios (unlimited dispersal, no dispersal, realistic dispersal, and realistic dispersal with long-distance dispersal events) and under four climate change scenarios.
Although simulations accounting for realistic dispersal limitations did significantly differ from those considering dispersal as unlimited or null in terms of projected future distribution, the unlimited dispersal simplification did nevertheless provide good approximations for species extinctions under more moderate climate change scenarios. Overall, simulations accounting for dispersal limitations produced, for our mountainous study area, results that were significantly closer to unlimited dispersal than to no dispersal. Finally, analysis of the temporal pattern of species extinctions over the entire 21st century revealed that important species extinctions for our study area might not occur before the 2080–2100 period, due to the possibility of a large number of species shifting their distribution to higher elevation.     

Climate change‐induced migration patterns and extinction risks of Theaceae species in China

Theaceae, an economically important angiosperm family, is widely distributed in tropical and subtropical forests in Asia. In China, Theaceae has particularly high abundances and endemism, comprising ~75% of the total genera and ~46% of the total species worldwide. Therefore, predicting the response of Theaceae species to climate change is vital. In this study, we collected distribution data for 200 wild Theaceae species in China, and predicted their distribution patterns under current and future climactic conditions by species distribution modeling (SDM). We revealed that Theaceae species richness is highest in southeastern China and on Hainan Island, reaching its highest value (137 species) in Fujian Province. According to the IUCN Red List criteria for assessing species threat levels under two dispersal assumptions (no dispersal and full dispersal), we evaluated the conservation status of all Theaceae species by calculating loss of suitable habitat under future climate scenarios. We predicted that nine additional species will become threatened due to climate change in the future; one species will be classified as critically endangered (CR), two as endangered (EN), and six as vulnerable (VU). Given their extinction risks associated with climate change, we recommended that these species be added to the Red List. Our investigation of migration patterns revealed regional differences in the number of emigrant, immigrant, and persistent species, indicating the need for targeted conservation strategies. Regions containing numerous emigrants are concentrated in Northern Taiwan and coastal regions of Zhejiang and Fujian provinces, while regions containing numerous immigrants include central Sichuan Province, the southeastern Tibet Autonomous Region, southwest Yunnan Province, northwest Sichuan Province, and the junction of Guangxi and Hunan provinces. Lastly, regions containing persistent species are widely distributed in southern China. Importantly, regions with high species turnover are located on the northern border of the entire Theaceae species distribution ranges owing to upwards migration; these regions are considered most sensitive to climate change and conservation planning should therefore be prioritized here. This study will contribute valuable information for reducing the negative impacts of climate change on Theaceae species, which will ultimately improve biodiversity conservation efficiency.     

Density and genetic relatedness increase dispersal distance in a subsocial organism

Although dispersal distance plays a major role in determining whether organisms will reach new habitats, empirical data on the environmental factors that affect dispersal distance are lacking. Population density and kin competition are two factors theorised to increase dispersal distance. Using the two‐spotted spider mite as a model species, we altered these two environmental conditions and measured the mean dispersal distance of individuals, as well as other attributes of the dispersal kernel. We find that both density and relatedness in the release patch increase dispersal distance. Relatedness, but not density, changes the shape of the dispersal kernel towards a more skewed and leptokurtic shape including a longer ‘fat‐tail’. This is the first experimental demonstration that kin competition can shape the whole distribution of dispersal distances in a population, and thus affect the geographical spread of dispersal phenotypes.     

Modeling bird species distribution change in fire prone Mediterranean landscapes: incorporating species dispersal and landscape dynamics

Species distribution models (SDMs) have traditionally been founded on the assumption that species distributions are in equilibrium with environmental conditions and that these species–environment relationships can be used to estimate species responses to environmental changes. Insight into the validity of this assumption can be obtained from comparing the performance of correlative species distribution models with more complex hybrid approaches, i.e. correlative and process‐based models that explicitly include ecological processes, thereby accounting for mismatches between habitat suitability and species occupancy patterns. Here we compared the ability of correlative SDMs and hybrid models, which can accommodate non‐equilibrium situations arising from dispersal constraints, to reproduce the distribution dynamics of the ortolan bunting Emberiza hortulana in highly dynamic, early successional, fire driven Mediterranean landscapes. Whereas, habitat availability was derived from a correlative statistical SDM, occupancy was modeled using a hybrid approach combining a grid‐based, spatially‐explicit population model that explicitly included bird dispersal with the correlative model. We compared species occupancy patterns under the equilibrium assumption and different scenarios of species dispersal capabilities. To evaluate the predictive capability of the different models, we used independent species data collected in areas affected to different degree by fires. In accordance with the view that disturbance leads to a disparity between the suitable habitat and the occupancy patterns of the ortolan bunting, our results indicated that hybrid modeling approaches were superior to correlative models in predicting species spatial dynamics. Furthermore, hybrid models that incorporated short dispersal distances were more likely to reproduce the observed changes in ortolan bunting distribution patterns, suggesting that dispersal plays a key role in limiting the colonization of recently burnt areas. We conclude that SDMs used in a dynamic context can be significantly improved by using combined hybrid modeling approaches that explicitly account for interactions between key ecological constraints such as dispersal and habitat suitability that drive species response to environmental changes.     

Dragonfly endemism in the Brazilian Amazon: competing hypotheses for biogeographical patterns

Many hypotheses have been proposed to explain the origin and maintenance of the Amazonian diversity with special place for the theory of isolation by rivers and a set of hypothesis related to contemporary environmental dissimilarity. We explore those hypotheses here using the biogeographic distributional patterns of dragonflies in interfluve areas of the Amazonian biome and also evaluate how differences among in dispersal capabilities between the Anisoptera and Zygoptera suborders may contribute to those patterns. We used distributional information of 392 odonate species in the Amazonian forest in a cladistic analysis of distributions and endemism and the estimated faunistic similarity among interfluves with the Sorensen index. The environmental similarity among interfluves was analysed by discriminant analysis based on eight environmental metrics. Different metrics for geographic distance (connectivity) among interfluves were evaluated and their relation to the other variables tested by the Mantel test. The number of endemic species was linearly correlated to the area of the interfluves. General endemism patterns showed consistent resemblance to those reported for vertebrates, especially the similarity among the Rond?nia and Inambari interfluves. Geographical distance has no predictive value for dragonflies distribution, but the environmental similarity is a good predictor of proportion of shared species. The low dispersal group (Zygoptera) presented more clear patterns of distribution and a lower proportion of shared species among different interfluves. The environmental similarity can be considered the determinant factor of the distribution of dragonflies, possibly due to environmental specificity evolved during a long history of some clades in this system. The low dispersal group (Zygoptera) retained more biogeographical information about possible historical factors that determine current distribution. Also, the transport of larvae by macrophyte banks, the lateral change of river courses, the reversal of the drainage basin, together with the capacity to disperse across rivers for some species may be explanations for the lack of effect of isolation by rivers, especially for Anisoptera.     

Natural selection, larval dispersal, and the geography of phenotype in the sea

Populations evolve generalist, specialist, and plastic strategies in response to environmental heterogeneity. Describing such within-species variation in phenotype and how it arises is central to understanding a variety of ecological and evolutionary topics. The literature on phenotypic differences among populations is highly biased; for every one article published on a marine species, at least 10 articles are published on a terrestrial species and eight focus on terrestrial plants. Here, I outline what we know from the marine literature about geographic variation in phenotype in the sea, with a principal focus on local adaptation. The theory of environmental "grain" predicts that the most likely evolutionary response (e.g., local adaptation, phenotypic plasticity, generalism, and balanced polymorphism) depends on the spatial scale of environmental variation relative to the distance that an organism disperses. Consistent with these predictions, phenotypic plasticity is stronger among invertebrates with geographically broad dispersal versus restricted dispersal (i.e., planktonic-dispersers versus direct-developers). However, contrary to predictions, the relative frequency, and spatial scale of local adaptation is not consistently greater among direct-developers relative to planktonic disperers. This indicates that the likelihood of local adaptation depends on other organismal or environmental traits. Two of the most vexing issues that remain include (1) predicting the extent to which barriers to dispersal are a cause versus consequence of phenotypic differentiation and (2) delineating the relative importance of evolutionary forces that favor or impede local adaptation. Understanding the mechanistic basis of the geography of phenotypic differences, or phenogeography, has gained recent momentum because of a need to predict impacts of global climatic change, anthropogenic disturbances, and dispersal of organisms to non-native habitats.     

Landscape genetics of an endangered lemur (Propithecus tattersalli) within its entire fragmented range

Habitat fragmentation may strongly reduce individuals’ dispersal among resource patches and hence influence population distribution and persistence. We studied the impact of landscape heterogeneity on the dispersal of the golden‐crowned sifaka (Propithecus tattersalli), an endangered social lemur species living in a restricted and highly fragmented landscape. We combined spatial analysis and population genetics methods to describe population units and identify the environmental factors which best predict the rates and patterns of genetic differentiation within and between populations. We used non‐invasive methods to genotype 230 individuals at 13 microsatellites in all the main forest fragments of its entire distribution area. Our analyses suggest that the Manankolana River and geographical distance are the primary structuring factors, while a national road crossing the region does not seem to impede gene flow. Altogether, our results are in agreement with a limited influence of forest habitat connectivity on gene flow patterns (except for North of the species’ range), suggesting that dispersal is still possible today among most forest patches for this species. Within forest patches, we find that dispersal is mainly among neighbouring social groups, hence confirming previous behavioural observations.     

Dispersal based climate change sensitivity scores for European species

Assessments of climate change impacts on species are needed for anticipating potential biodiversity losses. Climate change impacts on species are often simulated with climate envelope models, but most climate envelope models do not account for dispersal limitations. Most studies only consider two extreme (and unrealistic) dispersal options: no dispersal versus full dispersal. This study attempts to include dispersal limitation into the calculation of climate change sensitivity scores for a range of vertebrate and plant species. We calculate climate change sensitivity scores -expressed as an index- by using the 'spatial turnover' of a species under climate change, defined as the projected difference between current and future area occupied by a species within a region, and include a dispersal factor to account for dispersal limitations. We calculate climate sensitivity scores with three dispersal factors: d₀ (no dispersal), d₁ (full dispersal) and with an estimated value of d calculated directly from species specific dispersal data and literature estimates (d_e). We compared climate sensitivity scores across species groups and European bio-geographical regions in order to determine whether explicitly accounting for dispersal limitations causes significant differences in sensitivity for climate change. Our results show that the climate sensitivity scores calculated with d_e differ slightly from d₀ (no dispersal), but differ significantly from d₁ (full dispersal) for the less mobile species groups (amphibian, reptiles, plants). This indicates that assuming full dispersal significantly overestimates the future distribution in Europe under climate change for these species, whereas assuming no dispersal may slightly underestimates this. However, this conclusion could not be drawn for the more mobile birds and mammas: climate sensitivity scores calculated with d_e are approximately intermediate of those calculated with d₀ (no dispersal) and d₁ (full dispersal). This indicates that assuming either no or full dispersal results in poor estimates of the future distribution of these species in Europe under climate change, and that dispersal capacity should therefore always be considered when assessing climate change impacts on these species. Disaggregating climate sensitivity scores per European bio-geographical regions reveals that regional climate sensitivity scores are similar to the European level.     

Dietary guild composition and disaggregation of avian assemblages under climate change

Climate change is expected to cause geographic redistributions of species. To the extent that species within assemblages have different niche requirements, assemblages may no longer remain intact and dis‐ and reassemble at current or new geographic locations. We explored how climate change projected by 2100 may transform the world's avian assemblages (characterized at a 110 km spatial grain) by modeling environmental niche‐based changes to their dietary guild structure under 0, 500, and 2000 km‐dispersal distances. We examined guild structure changes at coarse (primary, high‐level, and mixed consumers) and fine (frugivores, nectarivores, insectivores, herbivores, granivores, scavengers, omnivores, and carnivores) ecological resolutions to determine whether or not geographic co‐occurrence patterns among guilds were associated with the magnitude to which guilds are functionally resolved. Dietary guilds vary considerably in their global geographic prevalence, and under broad‐scale niche‐based redistribution of species, these are projected to change very heterogeneously. A nondispersal assumption results in the smallest projected changes to guild assemblages, but with significant losses for some regions and guilds, such as South American insectivores. Longer dispersal distances are projected to cause greater degrees of disassembly, and lead to greater homogenization of guild composition, especially in northern Asia and Africa. This arises because projected range gains and losses result in geographically heterogeneous patterns of guild compensation. Projected decreases especially of primary and mixed consumers most often are compensated by increases in high‐level consumers, with increasing uncertainty about these outcomes as dispersal distance and degree of guild functional resolution increase. Further exploration into the consequences of these significant broad‐scale ecological functional changes at the community or ecosystem level should be increasingly on the agenda for conservation science.